Volcano Watch — Explosions from Kīlauea in the 16th to 18th centuries

How long do explosive periods last at Kīlauea? Early researchers concluded that the Keanakako`i Ash—the rocky deposits on the ground surface around Kīlauea caldera—was formed by many ex-plosions during several centuries ending in A.D. 1790, the known date of a large explosive eruption.

Workers in the 1970s to mid-1990s disagreed, concluding that the explosions took place for only a few years, perhaps entirely in 1790.

Evidence obtained in the past year forces us to return to the earlier interpretation. This change in thinking is important, for a prolonged period of episodic explosions presents different challenges to the community than one brief disastrous event.

Several layers of vegetation are buried within the Keanakako`i deposits on the windward side of the volcano. They signify enough time between some explosions for pukiawe, hapu`u, and other small plants to return. The layers, first noted during the 1920s and 1930s, are visible only in excavations. Pits north and south of Kīlauea Iki, and a cut along Kalanikoa Road in Volcano, expose three plant layers. In the 1920s, Thomas Jaggar described as many as six different layers during construction near the present Visitor Center.

The plant material is preserved as charcoal formed by fires possibly caused by eruptions. We have obtained more than 20 radiocarbon (carbon-14) ages from the different layers of charcoal. The ages range from about A.D. 1500 to 1790, consistent with about 300 years of explosive activity.

Hawaiians built several rock structures south of the caldera at two different times during the explosive period; one was excavated by Park Service archaeologists. Each structure was built on deposits of older explosions and partly covered by deposits of later explosions. The presence of the structures indicates lulls in explosive activity long enough for people to feel they could safely return to the area.

Other evidence argues for significant time between explosions. Water eroded deep channels into the deposits south of the caldera and washed older material away before later explosions deposited more debris. Wind removed sand from the explosion deposits in the Ka`u Desert and redeposited it in dunes. Six beds of explosive ash are interlayered with the dune sand. Two lava flows were erupted during lulls in explosive activity.

Many living o`hia lehua trees south of Kīlauea Iki, and even one near HVO, are too large to have grown entirely after A.D. 1790; one near Kīlauea Iki is probably about 350 years old. If the entire thickness of explosion debris (2 m [yards] or more) had been deposited in 1790, all trees would likely have been killed. This again suggests episodic accumulation of layers over some time.

Cultural evidence likewise suggests a long period of episodic explosive activity. In 1823, Hawaiians told Reverend William Ellis that "in earlier ages [Kīlauea] used to boil up, overflow its banks, and inundate the adjacent country [that is, the caldera had not yet formed]; but that for many kings' reigns past, it had kept below the level of the surrounding plain [the caldera had formed, keeping lava within it]?occasionally throwing up, with violent explosion, huge rocks or red-hot stones." Once "Pele was forced to her volcano" by angry Kamapua`a, who poured water to extinguish the fires; with great effort Pele "drank up the waters" and "finally succeeded in driving Kamapua`a into the sea, whither she followed him with thunder, lightning, and showers of large stones." This is a wonderful account of a steam-driven volcanic explosion before 1790.

The cultural and geologic evidence indicates that Kīlauea caldera formed in the late 15th century and was the source of sporadic explosive eruptions driven by heated groundwater or surface water "for many kings' reigns" before 1790. Will a similar prolonged explosive period follow the next deep caldera collapse?

Volcano Activity Update

Eruptive activity of Kīlauea Volcano continued unabated during the past week. Lava is erupting from Pu`u `O`o and flowing through a network of tubes from the vent to the sea. Short surface flows emanating from breakouts of the tube system were observed in the upper coastal flats. Lava is entering the ocean near Kamokuna and enlarging the bench. Occasional phreatic activity was observed at the entry. The public is reminded that the ocean-entry area is extremely hazardous, with explosions accompanying unpredictable collapses of the new land. The steam clouds are highly acidic and laced with glass particles.

There were no earthquakes reported felt during the week ending on August 5.